Oncocytic adrenocortical tumour presenting as an incidentaloma: a diagnostic challenge

Background

Oncocytic adrenocortical neoplasms (OAN) are a rare histopathological subtype of adrenocortical tumours, accounting for around 10% of adrenocortical neoplasms.1 An incidence of 0.72 per million population annually has been reported for oncocytic adrenocortical carcinoma (ACC).2 They are usually large, yellow-tan tumours with focal fatty areas. Histologically, they consist of a minimum of 90% oncocytes, which are epithelial cells with copious oeosinophilic and granular cytoplasm and numerous mitochondria, mostly arranged in a diffuse sheet-like pattern.3 4

Here we report a case of an OAN, first identified on imaging as an adrenal incidentaloma. Histology was consistent with malignancy when using the Lin-Weiss-Bisceglia (LWB) criteria but not when using the Helsinki scoring system. Despite meeting the criteria for malignancy when using the LWB system, this tumour exhibited an indolent course without any local recurrence or distant metastases throughout her follow-up of more than 10 years.

Case presentation

A woman in her 40s was initially referred to the endocrine clinic for further evaluation of a left adrenal incidentaloma. The left adrenal lesion was first identified on a CT scan of the abdomen and pelvis, which was performed as the patient developed lower abdominal pain 2 weeks following a total abdominal hysterectomy. It was first described as a 3.5 cm heterogeneous lesion in the left adrenal gland. The patient had no symptoms or signs of adrenal hormonal excess. Blood pressure was 110/70 mmHg, and the rest of the examination was unremarkable.

Investigations

Initial blood investigations included a normal renal profile with a serum potassium level of 4.8 mmol/L (3.5–5.1 mmol/L) and a serum sodium level of 135 mmol/L (135–145 mmol/L). Investigations for adrenal hormonal excess revealed normal consecutive 24-hour urinary cortisol levels with values of 95 nmol/24 hours, 83 nmol/24 hours, 104 nmol/24 hours and 94 nmol/24 hours, respectively (55–276 nmol/24 hours) and 540 nmol/24 hours, 491 nmol/24 hours, 449 nmol/24 hours and 549 nmol/24 hours (57.7–806.8 nmol/24 hours) during the first 5 years of follow-up. Subsequently, the patient however had two consecutively high 24-hour urinary cortisol levels of 1189 nmol/24 hours and 2263 nmol/24 hours (57.7–806.8 nmol/24 hours). Consequently, an overnight dexamethasone suppression test was carried out, which was normal as serum cortisol was successfully suppressed to 35 nmol/L (<50 nmol/L). Serum adrenocorticotropic hormone level was 11 pg/mL (10–48 pg/mL). Serial serum metanephrine and catecholamine levels were consistently within normal limits. An aldosterone/renin ratio was not done since the patient was normotensive and normokalaemic.

A repeat CT scan of the abdomen and pelvis was performed 5 months following the initial scan. The adrenal lesion was reported as a 3.5×3.4×5.6 cm inhomogeneous mass arising from the anterior limb of the left adrenal gland, with hypodense areas in the precontrast scan suggestive of lipid-rich foci. The findings were reported as being indicative of an adrenal myelolipoma. No delayed contrast-enhanced imaging was available at the time, and therefore, contrast washout could not be calculated.

A follow-up CT scan of the adrenals was done 1 year later. This was reported as showing no interval change in the size of the left adrenal lesion, which had an absolute enhancement washout of 60% and a relative washout of 40%, suggestive of an adenoma (figure 1). This is the lowest washout threshold for diagnosing an adenoma. A repeat CT scan of the adrenals 2 and 4 years later was also reported as not showing any interval change in size.

On CT adrenals 5 years following the original scan, an increase in the size of the left adrenal mass was described (dimensions now measuring 5.4×4.0×4.3 cm). In this scan, the density of the adrenal mass was calculated to be 20 Hounsfield units (HU) precontrast phase, 65 HU in the portal venous phase and 45 HU in the delayed phase, giving an absolute washout value of 44.4% (benign >60%) and a relative washout of 30.8% (benign >40%) (figures 2 and 3).

The increase in the size of the adrenal lesion prompted discussion of the initial CT scans with an experienced consultant radiologist, who calculated a density of 30 HU precontrast phase, 91 HU in the portal venous phase and 65 HU in the delayed phase, giving an absolute washout of 42.6% and a relative washout of 28.6% (figure 1). There was also a significant increase in the size of the adrenal mass from 3.4×3.5×4.9 cm in the original scans to 4.3×5.9×5.8 cm 5 years later (figures 2 and 3). The adrenal lesion was thus deemed to be indeterminate.

Differential diagnosis

The precontrast density of >10 HU effectively excluded a benign lipid-rich adenoma. Myelolipomas, although heterogenous, are characterised by the presence of macroscopic fat within the lesion often with a density of <−30 HU.5 6 Therefore, the major differential diagnoses included atypical benign lesions, such as lipid-poor adenomas, and malignant lesions. However, the low absolute and relative washouts made a diagnosis of a benign lipid-poor adenoma unlikely.7 A malignant lesion was thus thought to be more probable, with differentials including primary ACC, primary adrenal lymphoma (or more rarely adrenal angiosarcoma or leiomyosarcoma) and a secondary metastatic lesion. In view of the absence of evidence of malignant disease at other sites on cross-sectional imaging, a primary adrenal malignancy was thought to be more likely. Phaeochromocytoma, which could be benign or malignant, is another important differential to consider in an indeterminate adrenal neoplasm. However, the normal biochemical analysis for serum metanephrine levels would make a diagnosis of a phaeochromocytoma less likely. Other rare differential diagnoses of an indeterminate adrenal mass include benign haemangiomas, ganglioneuromas and OAN, the latter of which can be benign or malignant.6 8–11

Treatment

Due to the increase in size and imaging features that were not consistent with a benign adenoma, the patient was referred for surgery and underwent laparoscopic left adrenalectomy almost 6 years after her first review at the endocrine clinic.

Outcome and follow-up

Histology revealed an encapsulated mass measuring 7.5×4.0×4.2 cm attached to a portion of an unremarkable adrenal gland measuring 5.5×3.5×0.5 cm. It weighed 92 g. On sectioning, the tumour had a variegated, yellow myxoid surface with foci of haemorrhagic and cystic degeneration. The capsule appeared to be intact, and the tumour was continuous with the adrenal gland. Microscopically, a diffusely oncocytic tumour was noted to occupy most of the adrenal cortex, obliterating the adrenal medulla. The tumour displayed oncocytic cells with well-demarcated cell borders and predominantly sheet-like architecture, but with occasional nested architecture (figure 4). Severe nuclear pleomorphism and multinucleated forms were noted. Numerous atypical mitoses were identified focally with no areas of necrosis. Mitoses averaged 7/50 high-power fields (HPF). There was evidence of vascular invasion but no perineural invasion. The tumour exhibited diffuse melan-A expression, patchy inhibin and cytokeratin expression and no expression of S-100. Ki-67 was calculated to be between 3% and 5% (figure 4). It was reported as consistent with an oncocytic ACC.

A staging CT scan was thus performed which showed no evidence of metastases. This was confirmed on an ¹⁸F-fluorodeoxyglucose positron emission tomography-CT (¹⁸F-FDG PET-CT) scan. The patient was reviewed by an oncologist, and a decision was taken to observe the patient. No adjuvant therapy was administered. On serial imaging to date, the patient has remained well with no evidence of local disease recurrence or metastases.